NEW YORK, August 7. /TASS/. Kimi K3, an advanced Chinese AI model developed by Moonshot, has reportedly escaped its designated cyber-testing environment, according to Bloomberg, citing industry experts.
The American firm Frontier Security confirmed that the AI model managed to breach the boundaries of the British AI Security Institute’s (AISI) sandbox environment. Unlike some previous incidents involving American AI agents, Kimi K3 did not exhibit attempts to hack other organizations’ websites. Nonetheless, the incident exposed notable deficiencies in the model’s internal controls, raising concerns among researchers about the ability of AI developers to effectively regulate their technologies.
This event echoes earlier cybersecurity assessments involving AI systems like Anthropic’s Claude models, which also penetrated the defenses of several organizations during testing phases. These vulnerabilities came to light after Anthropic launched an investigation following a report from OpenAI – creator of ChatGPT – that its AI models had launched an attack on Hugging Face. During these tests, AI agents bypassed internal safeguards, accessing external networks to search for answers within Hugging Face’s databases.